f01introl_2013.pdf
TRANSCRIPT
![Page 1: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/1.jpg)
FRTN10 Multivariable Control — Lecture 1
Anders Rantzer
Automatic Control LTH, Lund University
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 2: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/2.jpg)
Todays lecture
Introduction/examples
Overview of course + feedback/feedforward
Review linear systems
Review of time-domain modelsReview of frequency-domain modelsNorm of signalsGain of systems
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 3: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/3.jpg)
Many actuators and measurements
Example: Control of Large Deformable Telescope Mirror
Large number of sensors and actuators (500-3000)
Computational limitations (1kHz)
Tolerance ( 1 nano-meter
Control accuracy crucial for telescope performance!
See more at e.g., http://www.tmt.org/
http://www.astro.lu.se/∼torben/euro50/index.htmlAutomatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 4: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/4.jpg)
Example: Rollover protection needed
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 5: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/5.jpg)
Rollover Control
ControllerControl
AllocatorActuator Dynamics
Controller Plant
v u yr
U
V
r
d
af
ar
b
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 6: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/6.jpg)
Car dynamics
U
V
r
δ
Vehicle
brake forces {ui}
steering angle (δ )
lateral velocity (V )
yaw rate (r)
State space model
[
V
r
]
= A
[
V
r
]
+
[
0
b1
]
(u1 + u2 − u3 − u4) +
[
b2b3
]
δ
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 7: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/7.jpg)
Fredrik Arp (Volvo) on Environmental Issues
[Sydsvenskan 2007]:
“Genom effektivisering av de konventionella bensin- och
dieselmotorerna kan vi hämta hem en besparing på 20 procent
i emissioner och bränsleekonomi de närmaste fem-sex åren”
Med andra ord: Bättre reglering gör skillnad!
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 8: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/8.jpg)
The DVD reader tracking problem
Pit
Track
0.74 µm
3.5 m/s speed along track
0.022 µm tracking tolerance
100 µm deviations at 23 Hz due
to asymmetric discs
DVD Digital Versatile Disc, 4.7 Gb
CD Compact Disc, 650 Mb, mostly audio and software
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 9: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/9.jpg)
The DVD pick-up head
Radial electromagnet
Focus electromagnet
Springs
Light detectors
Laser
A B
C D
Tracks
Lens
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 10: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/10.jpg)
Input-output diagram for DVD control
PUH & Disk
vertical force
radial force
Focus error
Radial error
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 11: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/11.jpg)
The DVD reader in our lab
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 12: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/12.jpg)
Control problem
r
+
C P
−1
u z
y n
w
Given the system P and measurement signals y, determine the
control signals u such that the control objective z follows the
reference r as “close as possible” despite disturbances w,
measurement errors n (noise etc.) and uncertainties of the real
process.
For closed-loop ctrl =[ determine controller C.
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 13: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/13.jpg)
Mid-ranging control
Example: Radial control of pich-up-head of DVD-player
Radial electromagnet
Focus electromagnet
Springs
Light detectors
Laser
A B
C D
Tracks
Lens
Pick−up headSledge
Disk
The pick-up-head has two electromagnets for fast positioning of the lens
(left). Larger radial movements are taken care of by the sledge (right).
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 14: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/14.jpg)
DVD in the course
Focus control and tracking control lectured as a design
example (Case study lecture 5)
What do we learn?
Challenging design excercises
Respect fundamental limitations
Sampling frequency critical
The use of observers
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 15: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/15.jpg)
The design process
Experiment
Implementation
Synthesis
Analysis
Matematical modeland
specification
Idea/Purpose
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 16: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/16.jpg)
Contents of the course
L1-L5 Specifications, models and loop-shaping by hand
L6-L8 Limitations on achievable performance
L9-L11 Controller optimization: Analytic approach
L12-L14 Controller optimization: Numerical approach
Experiment
Implementation
Synthesis
Analysis
Matematical modeland
specification
Idea/Purpose
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 17: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/17.jpg)
Course home page
http://www.control.lth.se/Education/EngineeringProgram/FRTN10.html
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 18: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/18.jpg)
Literature
T. Glad and L. Ljung:
Svensk utgåva: Reglerteori – Flervariabla och
olinjära metoder , 2nd ed Studentlitteratur, 2004English translation: Control Theory – Multivariable
and Nonlinear Methods, Taylor and Francis
Lecture Slides/Notes on the web
Exercise problems with solutions on the web
Laboratory PMs
Swedish-English control dictionary on homepage
KFS sells the book
Course web page:
http://www.control.lth.se/course/FRTN10
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 19: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/19.jpg)
Lectures
The lectures (30 hours) are given as follows:
Mondays Sep 2, 9, 16, 23 and Oct 14 8.15 in MH:A except today!Wednesdays Sep 4, 11, 18 and Oct 2, 9, 16 8.15 in M:BThursdays Sep 5, 19, 26 and Oct 3 15.15 in M:B
All course material is in English.
The lectures are given by
Anders Rantzer and Per Hagander
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 20: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/20.jpg)
Exercise sessions and TAs
The exercises (28 hours) are taught according to the schedule
First session Monday 13–15 Monday 15–17 lab A and BSecond session Thursday 13–15 Friday 13–15 lab A and B
They are all held in the department laboratory on the bottom
floor in the south end of the Mechanical Engineering building.
Fredrik Magnusson Jerker Nordh Josefin Berner Ola Johnsson
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 21: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/21.jpg)
Laboratory experiments
The three laboratory experiments are mandatory.
Sign-up lists are posted on the web at least one week before
the first laboratory experiment. The lists close one day before
the first session.
The Laboratory PMs are available at the course homepage.
Before the lab sessions some home assignments have to bedone. No reports after the labs.
Lab Week Booking Starts Responsible ContentLab 1 w 38-39 Sep 11 Josefin Berner Flex-servoLab 2 w 41 Sep 23 Josefin Berner Quad-tankLab 3 w 42 Oct 7 Fredrik Magnusson Crane
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 22: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/22.jpg)
Laboratory experiments
The three laboratory experiments are mandatory.
Sign-up lists are posted on the web at least one week before
the first laboratory experiment. The lists close one day before
the first session.
The Laboratory PMs are available at the course homepage.
Before the lab sessions some home assignments have to bedone. No reports after the labs.
Lab Week Booking Starts Responsible ContentLab 1 w 38-39 Sep 11 Josefin Berner Flex-servoLab 2 w 41 Sep 23 Josefin Berner Quad-tankLab 3 w 42 Oct 7 Fredrik Magnusson Crane
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 23: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/23.jpg)
Exam
The exam (5 hours) will be given
Wednesday Oct 23.
Lecture notes and text book are allowed, but no exercises
material or extra hand-written notes.
Next time January 8, 2013 ( pre-register on web
http://www.control.lth.se/Education/EngineeringProgram ).
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 24: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/24.jpg)
Use of computers in the course
Use personal student-account or a common course
account
Matlab in exercises and laboratories (!!)
Web page:
http://www.control.lth.se/Education/EngineeringProgram/FRTN10
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 25: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/25.jpg)
Feedback is important
For each course LTH use the following feedback mechanisms
CEQ (reporting / longer time scale)
Student representatives (fast feedback)
Election of student representative ("kursombud")
Email [email protected]
Help us close the loop for better performance!
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 26: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/26.jpg)
Registration
You must register for the course by signing the form
available upfront during the break (will be passed around also
during the 2nd hour).
If your name is not in the form please fill in an empty row.
LADOK registration will be done immediately.
If you decide to abort/skip the course within three weeks from
today you should inform me and then the LADOK registration
will be removed.
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 27: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/27.jpg)
Course Outline
L1-L5 Specifications, models and loop-shaping by hand
1 Introduction and system representations2 Stability and robustness3 Specifications and disturbance models4 Control synthesis in frequency domain5 Case study
L6-L8 Limitations on achievable performance
L9-L11 Controller optimization: Analytic approach
L12-L14 Controller optimization: Numerical approach
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 28: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/28.jpg)
Lecture 1
Description of linear systems (different representations)
Review of time-domain models
Review of frequency-domain models
Norm of signals
Gain of systems
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 29: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/29.jpg)
State Space Equations
State-space and time-solution
{
x = Ax + Bu
y= Cx + Du
y(t) = CeAtx(0) +
∫ t
0
CeA(t−τ )Bu(τ )dτ + Du(t)
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 30: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/30.jpg)
Example
x1 = −x1 + 2x2 + u1 + u2 − u3
x2 = −5x2 + 3u2 + u3
y1 = x1 + x2 + u3
y2 = 4x2 + 7u1
How many states, inputs and outputs?
x = Ax + Bu
y= Cx + Du
[
x1x2
]
=
[
∗ ∗
∗ ∗
] [
x1x2
]
+
[
∗ ∗ ∗
∗ ∗ ∗
]
u1u2u3
[
y1y2
]
=
[
∗ ∗
∗ ∗
] [
x1x2
]
+
[
∗ ∗ ∗
∗ ∗ ∗
]
u1u2u3
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 31: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/31.jpg)
Example
x1 = −x1 + 2x2 + u1 + u2 − u3
x2 = −5x2 + 3u2 + u3
y1 = x1 + x2 + u3
y2 = 4x2 + 7u1
How many states, inputs and outputs?
x = Ax + Bu
y= Cx + Du
[
x1x2
]
=
[
∗ ∗
∗ ∗
] [
x1x2
]
+
[
∗ ∗ ∗
∗ ∗ ∗
]
u1u2u3
[
y1y2
]
=
[
∗ ∗
∗ ∗
] [
x1x2
]
+
[
∗ ∗ ∗
∗ ∗ ∗
]
u1u2u3
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 32: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/32.jpg)
Example
x1 = −x1 + 2x2 + u1 + u2 − u3
x2 = −5x2 + 3u2 + u3
y1 = x1 + x2 + u3
y2 = 4x2 + 7u1
How many states, inputs and outputs?
x = Ax + Bu
y= Cx + Du
[
x1x2
]
=
[
∗ ∗
∗ ∗
] [
x1x2
]
+
[
∗ ∗ ∗
∗ ∗ ∗
]
u1u2u3
[
y1y2
]
=
[
∗ ∗
∗ ∗
] [
x1x2
]
+
[
∗ ∗ ∗
∗ ∗ ∗
]
u1u2u3
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 33: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/33.jpg)
Example
x1 = −x1 + 2x2 + u1 + u2 − u3
x2 = −5x2 + 3u2 + u3
y1 = x1 + x2 + u3
y2 = 4x2 + 7u1
[
x1x2
]
=
[
−1 2
0 −5
] [
x1x2
]
+
[
1 1 −10 3 1
]
u1u2u3
[
y1y2
]
=
[
1 1
0 4
] [
x1x2
]
+
[
0 0 1
7 0 0
]
u1u2u3
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 34: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/34.jpg)
State space form cont’d
Exampel:
2nd order differential equation
y+ 3y+ 2y= 5u
Write on state space form.
How to chose states?
What if derivatives of input signal appears?
Superposition
Canonical forms
Collection of formulae
...
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 35: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/35.jpg)
State space form cont’d
Exampel:
2nd order differential equation
y+ 3y+ 2y= 5u
Write on state space form.
How to chose states?
What if derivatives of input signal appears?
Superposition
Canonical forms
Collection of formulae
...
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 36: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/36.jpg)
Change of coordinates
{
x = Ax + Bu
y= Cx + Du
Change of coordinates
z = Tx
{
z = T x = T(Ax + Bu) = T(AT−1z+ Bu) = TAT−1z+ TBu
y= Cx + Du = CT−1z+ Du
Note: There are many different state-space representations for
the same transfer function and system!
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 37: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/37.jpg)
Change of coordinates
{
x = Ax + Bu
y= Cx + Du
Change of coordinates
z = Tx
{
z = T x = T(Ax + Bu) = T(AT−1z+ Bu) = TAT−1z+ TBu
y= Cx + Du = CT−1z+ Du
Note: There are many different state-space representations for
the same transfer function and system!
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 38: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/38.jpg)
Impulse response
-1 0 1 2 3 4 5 6 7 8 9 10
0
0.2
0.4
0.6
0.8
1
-1 0 1 2 3 4 5 6 7 8 9 10-0.1
0
0.1
0.2
0.3
0.4
0.5
y(t)=�(t)
u(t)=
δ(t)
Common experiment in medicin and biology
�(t) =
∫ t
0
CeA(t−τ )Bδ (τ )dτ + Dδ (t) = CeAtB + Dδ (t)
y(t) =
∫ t
0
�(t− τ )u(τ )dτ = [� ∗ u](t)
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 39: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/39.jpg)
Step response
-1 0 1 2 3 4 5 6 7 8 9 10
0
0.2
0.4
0.6
0.8
1
-1 0 1 2 3 4 5 6 7 8 9 10
0
0.2
0.4
0.6
0.8
1
y(t)
u(t)
Common experiment in process industry
y(t) =
∫ t
0
�(t− τ )u(τ )dτ
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 40: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/40.jpg)
Frequency response
0 2 4 6 8 10 12 14 16 18 20-0.2
-0.1
0
0.1
0.2
0.3
y
0 2 4 6 8 10 12 14 16 18 20-1.5
-1
-0.5
0
0.5
1
1.5
u
The transfer function G(s) is the Laplace transform of the
impulse response G = L�. The input u(t) = sinω t gives
y(t) =
∫ t
0
�(τ )u(t− τ )dτ = Im
[∫ t
0
�(τ )e−iωτdτ ⋅ eiω t]
[t→∞] = Im(
G(iω )eiω t)
= pG(iω )p sin(
ω t+ argG(iω ))
After a transient, also the output becomes sinusoidal
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 41: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/41.jpg)
The Nyquist Diagram
-0.4 -0.2 0 0.2 0.4 0.6 0.8 1 1.2-1
-0.8
-0.6
-0.4
-0.2
0
0.2
argG(iω )
pG(iω )p
Im G(iω )
Re G(iω )
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 42: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/42.jpg)
Asymptotic formulas for first order system
-0.2 0 0.2 0.4 0.6 0.8 1 1.2-1
-0.8
-0.6
-0.4
-0.2
0
0.2
G(s) =1
s+ 1
G(iω ) =1
iω + 1=1− iω
ω 2 + 1
Small ω : G(iω ) ( 1
Large ω : G(iω ) (1
ω 2− i1
ω
Matlab:
» s=tf(’s’);
» G=1/(s+1);
» nyquist(G)
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 43: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/43.jpg)
Asymptotic formulas for first order system
-0.2 0 0.2 0.4 0.6 0.8 1 1.2-1
-0.8
-0.6
-0.4
-0.2
0
0.2
G(s) =1
s+ 1
G(iω ) =1
iω + 1=1− iω
ω 2 + 1
Small ω : G(iω ) ( 1
Large ω : G(iω ) (1
ω 2− i1
ω
Matlab:
» s=tf(’s’);
» G=1/(s+1);
» nyquist(G)
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 44: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/44.jpg)
The Bode Diagram
10-1
100
101
102
103
10-4
10-2
100
A
B
10-1
100
101
102
103
-180
-160
-140
-120
-100
-80
-60
-40
Frekvens [rad/s]
Am
plit
ud
eP
ha
se
G = G1G2G3
{
log pGp = log pG1p + log pG2p + log pG3p
argG = argG1 + argG2 + argG3
Each new factor enter additively!
Hint: Set matlab-scales
» ctrlpref
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 45: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/45.jpg)
The Bode Diagram
10-1
100
101
102
103
10-4
10-2
100
A
B
10-1
100
101
102
103
-180
-160
-140
-120
-100
-80
-60
-40
Frekvens [rad/s]
Am
plit
ud
eP
ha
se
G = G1G2G3
{
log pGp = log pG1p + log pG2p + log pG3p
argG = argG1 + argG2 + argG3
Each new factor enter additively!
Hint: Set matlab-scales
» ctrlpref
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 46: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/46.jpg)
The L2-norm of a signal
For y(t) ∈ Rn the “L2-norm”
qyq2 :=
√
∫ ∞
0
py(t)p2dt is equal to
√
1
2π
∫ ∞
−∞pLy(iω )p2dω
The equality is known as Parseval’s formula
The L2-gain of a system For a system S with input u and
output S(u), the L2-gain is defined as
qSq := supu
qS(u)q2quq2
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 47: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/47.jpg)
Miniproblem
What are the gains of the following systems?
1. y(t) = −u(t) (a sign shift)
2. y(t) = u(t− T) (a time delay)
3. y(t) =
∫ t
0
u(τ )dτ (an integrator)
4. y(t) =
∫ t
0
e−(t−τ )u(τ )dτ (a first order filter)
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 48: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/48.jpg)
The L2-gain from frequency data
Consider a stable system S with input u and output S(u) having
the transfer function G(s). Then, the system gain
qSq := supu
qS(u)q2quq2
is equal to qGq∞ := supωpG(iω )p
Proof. Let y= S(u). Then
qyq2 =1
2π
∫ ∞
−∞
pLy(iω )p2dω =1
2π
∫ ∞
−∞
pG(iω )p2 ⋅ pLu(iω )p2dω ≤ qGq2∞quq2
The inequality is arbitrarily tight when u(t) is a sinusoid near
the maximizing frequency.
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 49: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/49.jpg)
W. Wright at Western Society of Engineers 1901
“Men already know how to construct wings or airplanes, which
when driven through the air at sufficient speed, will not only
sustain the weight of the wings themselves, but also that of the
engine, and of the engineer as well. Men also know how to
build engines and screws of sufficient lightness and power to
drive these planes at sustaining speed ... Inability to balance
and steer still confronts students of the flying problem. ...
When this one feature has been worked out, the age of flying
will have arrived, for all other difficulties are of minor
importance.”
Wright was right!
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 50: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/50.jpg)
W. Wright at Western Society of Engineers 1901
“Men already know how to construct wings or airplanes, which
when driven through the air at sufficient speed, will not only
sustain the weight of the wings themselves, but also that of the
engine, and of the engineer as well. Men also know how to
build engines and screws of sufficient lightness and power to
drive these planes at sustaining speed ... Inability to balance
and steer still confronts students of the flying problem. ...
When this one feature has been worked out, the age of flying
will have arrived, for all other difficulties are of minor
importance.”
Wright was right!
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 51: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/51.jpg)
Smart Grid Gotland
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 52: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/52.jpg)
SURF - exchange program with Caltech
Example: DARPA Grand Challenge and Team Caltech
Autonomously Los Angeles to Las Vegas in < 10 h in 2004
Lund students in SURF programme at Caltech every year
http://www.control.lth.se/SURF
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1
![Page 53: f01Introl_2013.pdf](https://reader034.vdocuments.us/reader034/viewer/2022051518/55cf98e5550346d0339a509a/html5/thumbnails/53.jpg)
Course Outline
L1-L5 Specifications, models and loop-shaping by hand
1 Introduction and system representations2 Stability and robustness3 Disturbance models4 Control synthesis in frequency domain5 Case study
L6-L8 Limitations on achievable performance
L9-L11 Controller optimization: Analytic approach
L12-L14 Controller optimization: Numerical approach
Automatic Control LTH, 2012 FRTN10 Multivariable Control — Lecture 1